|
Abstract
|
Background
The Multidimensional Driving Style Inventory (MDSI, Taubman-Ben-Ari et al., 2004) is a widely used instrument that assesses private car drivers' driving styles. However, a key validity check for the MDSI is whether the measurement of driving style is associated with drivers' crash liability.
Methods
A one-year follow-up study was conducted, and one thousand bus drivers were randomly recruited from three cities (Hangzhou, Beijing and Bazhong) in China. The participants were asked to complete the Chinese version of the MDSI and a demographic questionnaire, and they needed to report their traffic violation history 12 months after the survey. Violation history data only included traffic violations and crashes that were caused by the study participants rather than other road users. A confirmatory factor analysis (CFA) was conducted to assess the factorial structure of the MDSI. The relationships between the MDSI and traffic violation history over the previous 12 months and driving patterns over the 12 months immediately following the survey were investigated.
Results
CFA results show that the model fitness of the MDSI for bus drivers is acceptable. Both drivers' traffic violations and crashes over the previous 12 months and driving patterns over the following 12 months were positively correlated with dissociative, risky and angry styles and negatively correlated with a careful style. Cluster analysis results show three clusters, with one unsafe driver group characterized by higher scores on dissociative, anxious, risky and angry styles and lower scores on careful driving styles. The number of drivers with traffic violations or crashes in the unsafe cluster is significantly higher than those in the other two clusters. More importantly, drivers in the unsafe cluster experience more traffic violations and crashes over the following 12 months than the other two driver clusters.
Conclusion
The findings of the present study not only provide a reliable and valid instrument for assessing bus drivers' driving styles but also show valid evidence for the predictive validity of the MDSI in measuring bus drivers' crash involvement.
Language: en |